摘要
针对BP神经网络收敛速度慢,易于陷入局部极小点的问题,将蚁群算法引入BP神经网络的优化训练,建立了基于该算法的BP神经网络训练模型,并应用于电机转子故障诊断。结果表明,用蚁群算法训练神经网络具有较高的故障诊断精度,收敛性好,可以有效快速定位电机转子故障,提高诊断的效率和质量。
To overcome shortcomings of BP neural network,ACO algorithm is introduced into optimization training of BP neural network.The fault diagnosis system of neural network is established.ACO algorithm is used to train a neural network for fault diagnosis of motor.The diagnostic results based on ACO algorithm are compared with those of BP algorithm.Comparisions of network training show that fault diagnosis system based on ACO algorithm has a better identification probability of faults for multi-fault symptoms.The results show that ACO algorithm has faster convergence rate,higher accuracy and searching efficiency.
出处
《北京信息科技大学学报（自然科学版）》
2010年第2期45-48,共4页
Journal of Beijing Information Science and Technology University
基金
国家自然科学基金(50975020)
北京市属高等学校人才强教计划资助项目(PHR20090518)
关键词
蚁群算法
BP神经网络
故障诊断
ant colony optimization（ACO）
BP neural network
fault diagnosis
